Expected Label Value Computation for Atlas-Based Image Segmentation.

Expected Label Value Computation for Atlas-Based Image Segmentation. Proc IEEE Int Symp Biomed Imaging. 2019 Apr;2019:334-338 Authors: Aganj I, Fischl B Abstract The use of multiple atlases is common in medical image segmentation. This typically requires deformable registration of the atlases (or the average atlas) to the new image, which is computationally expensive and susceptible to entrapment in local optima. We propose to instead consider the probability of all possible transformations and compute the expected label value (ELV), thereby not relying merely on the transformation resulting from the registration. Moreover, we do so without actually performing deformable registration, thus avoiding the associated computational costs. We evaluate our ELV computation approach by applying it to liver segmentation on a dataset of computed tomography (CT) images. PMID: 31341547 [PubMed]
Source: Proceedings - International Symposium on Biomedical Imaging - Category: Radiology Tags: Proc IEEE Int Symp Biomed Imaging Source Type: research